I made LLM context editable: a graph where the wires are the prompt

Short summary

ThoughtDAG is an open-source MIT-licensed prototype that represents LLM conversations as a directed acyclic graph on an infinite canvas, where edges determine which upstream nodes are included in the next model request. Users can branch, merge, and prune context explicitly instead of relying on a linear transcript. It runs locally in the browser with IndexedDB storage and connects to Ollama or any OpenAI-compatible endpoint.

• •LLM context represented as a DAG where edges define prompt composition
• •Supports branching, merging, and pruning of conversation context
• •MIT licensed, browser-local storage, connects to Ollama or OpenAI-compatible endpoints

Generated with AI, which can make mistakes.